HIV-1 has infected 37.8 million persons in the world and 1.2 million are living with HIV-1 in the U.S. New York has the most AIDS cases at 172,370 (18% of the U.S. total) and the highest case rate at 32.7/10A5. South Africa, which has <2% of the world's population, is home to 30% of all the people living with HIV/AIDS in the world (5.3 million). HAART reduces plasma viral load below the limits of detection in most patients, increases CD4+ counts, and reduces the incidence of opportunistic infections. Pulmonary infectious complications are still recognized: bacterial pneumonia, tuberculosis, CMV and PCP. We hypothesize that bacterial pneumonia occurs in the HAART era, albeit at a lower incidence, and will enhance local HIV-1 replication and mutation. AIM 1 will enroll 200 HIV-1+ patients each at New York and Cape Town, collect questionnaires, blood, sputum, PFT, and follow them prospectively for 4 annual visits. AIM 2 will study 36 HIV-1+ patients with bacterial pneumonia and 72 HIV-1 + controls at each site with bronchoalveolar lavage (BAL) of involved/uninvolved lobes. HIV-1 will be quantitated in involved and uninvolved lobes and plasma with the expectation that inflammation with PMNs may drive macrophage HIV-1 transcription increasing HIV- 1 RNA viral load, RT-PCR in macrophages and increased envelope V3 loop mutations and drug resistance mutations in involved lobes. We will characterize the cells in lung and blood with surface markers using flow cytometry. AIM 3 will evaluate macrophage-PMN interaction to determine the mechanisms of Reactive Oxidant Species damage that may cause mutations using in vitro experiments and BAL cells. We will evaluate Treg suppressive function in Cape Town to determine effects of HAART and HIV-1 on Tregs followed by changes caused by bacterial pneumonia in involved/uninvolved lobe compartments. This research will provide guidance on HAART therapy in HIV-1+ patients, particularly when the CD4+ level is preserved >250 cells/ml and whether HAART can prevent bacterial pneumonia, or alter its clinical features. We will also determine whether bacterial pneumonia predisposes to HAART failure, and if pneumonia in patients with CD4+ should be started on HAART because the focal lung inflammation increases HIV-1 replication and mutations. In this prospective study, we will also evaluate latent tuberculosis and other respiratory infections to determine their influence on the course of HIV-1 infection. We will use high-speed FACS cell sorting to separate macrophages, lymphocytes, and neitrophils for mechanistic studies. We will freeze and store at -70 degrees plasma, serum, PBMC (RNA, DNA, protein), BAL, from involved/uninvolved lung segments, and sputum for collaborative studies with other NHLBI grantees.